Nextstrain-augur

Major Changes

* filter: Skip metadata records with ambiguous month information in the `date` column when grouping by month instead of randomly generating month values for those records. This change alters the behavior of the `filter` command for metadata with ambiguous month values. For these data, consider using `--group-by year` instead of `--group-by year month`. [761][] (huddlej)

Features

* filter: When grouping by year or month, report the number of strains skipped due to ambiguous year and month both in the summary report at the end of filtering and in the `--output-log` contents [761][] (huddlej)

[761]: https://github.com/nextstrain/augur/pull/761

Bug Fixes

* filter: Fix parsing of missing data in metadata [758][] (huddlej)
* filter: Fix probabilistic sampling with small values [759][] (huddlej)

[758]: https://github.com/nextstrain/augur/pull/758
[759]: https://github.com/nextstrain/augur/pull/759

Features

* export: Add support for custom legend and color scale specifications in Auspice config files [727][] (jameshadfield)
* utils: Add support for compressed strain name files (e.g., "include.txt.gz") [730][] (benjaminotter)
* filter: Rewrite internal logic to use pandas DataFrames ([743][]), define filters and subsampling logic as individual functions ([745][] and [746][]), and iterate through chunks of metadata instead of loading all records into memory at once ([750][]) (tsibley, huddlej)

Bug Fixes

* distance: Change numeric type of distance output to float [729][] (benjaminotter)
* filter: Disable probabilistic sampling when users provide `--sequences-per-group` [737][] (benjaminotter)
* export: Provide correct missing file error messages for metadata and node data JSON inputs [752][] (benjaminotter)

[727]: https://github.com/nextstrain/augur/pull/727
[729]: https://github.com/nextstrain/augur/pull/729
[730]: https://github.com/nextstrain/augur/pull/730
[737]: https://github.com/nextstrain/augur/pull/737
[743]: https://github.com/nextstrain/augur/pull/743
[745]: https://github.com/nextstrain/augur/pull/745
[746]: https://github.com/nextstrain/augur/pull/746
[750]: https://github.com/nextstrain/augur/pull/750
[752]: https://github.com/nextstrain/augur/pull/752

Major Changes

* filter: Date bounds (`--min-date` and `--max-date`) are now inclusive instead of exclusive such that records matching the given dates will pass date filters [708][] (benjaminotter)

Bug Fixes

* refine: Recommend an alternate action when skyline optimization fails [712][] (huddlej)

Features

* distance: Count insertion/deletion events once in pairwise distances [698][] (huddlej, benjaminotter)
* distance: Optionally ignore specific list of characters defined in a distance map's top-level `ignored_characters` list [707][] (benjaminotter)
* filter: Allow `--subsample-max-sequences` without `--group-by` [710][] (benjaminotter)
* tree: Prefer `iqtree2` binary over `iqtree` when possible [711][] (benjaminotter)

[698]: https://github.com/nextstrain/augur/pull/698
[707]: https://github.com/nextstrain/augur/pull/707
[708]: https://github.com/nextstrain/augur/pull/708
[710]: https://github.com/nextstrain/augur/pull/710
[711]: https://github.com/nextstrain/augur/pull/711
[712]: https://github.com/nextstrain/augur/pull/712

Bug Fixes

* filter: Clarify how the `--priority` input affects subsampling in the command line help documentation [695][]
* tests: Clean up outputs created by tests [703][], ignore log files [701][], and remove unnecessary Conda environment file [702][]

Features

* io: Add new `io` module with `open_file`, `read_sequences`, and `write_sequences` functions that support compressed inputs and outputs [652][]
* parse, index, filter, mask: Add support for compressed inputs/outputs [652][]
* export v2: Add optional `data_provenance` field to auspice JSON output for better provenance reporting in Auspice [705][]

[652]: https://github.com/nextstrain/augur/pull/652
[695]: https://github.com/nextstrain/augur/pull/695
[701]: https://github.com/nextstrain/augur/pull/701
[702]: https://github.com/nextstrain/augur/pull/702
[703]: https://github.com/nextstrain/augur/pull/703
[705]: https://github.com/nextstrain/augur/pull/705

Bug Fixes

* ancestral: Mask positions that are ambiguous in all tip sequences before inferring ancestral sequence states, to avoid assigning arbitrary ancestral values based on rounding errors [682][]
* titers: Add missing `kwargs` attribute to `TiterCollection` class [690][]

Documentation

* Update API documentation to include newer Python modules and the `index` subcommand [687][]
* Remove Zika and TB tutorials in favor of copies in docs.nextstrain.org [689][]

Features

* filter: Enable filtering by metadata only such that sequence inputs/outputs are optional and metadata/strain list outputs are now possible [679][]
* filter: Enable extraction of sequences from multiple lists of strains with a new `--exclude-all` flag and support for multiple inputs to the `--include` argument [679][]

[679]: https://github.com/nextstrain/augur/pull/679
[682]: https://github.com/nextstrain/augur/pull/682
[687]: https://github.com/nextstrain/augur/pull/687
[689]: https://github.com/nextstrain/augur/pull/689
[690]: https://github.com/nextstrain/augur/pull/690